Constraints in multi-objective optimization of land use allocation – Repair or penalize?. (August 2019)

• Record Type: Journal Article
• Authors: Strauch, Michael; Cord, Anna F.; Pätzold, Carola; Lautenbach, Sven; Kaim, Andrea; Schweitzer, Christian; Seppelt, Ralf; Volk, Martin
• Is Part Of: Environmental modelling & software. Volume 118(2019)
• Page Start: 241
• Page End: 251
• Publication Date: 2019-08
• Subjects: Land use allocation -- Ecosystem services -- Biodiversity -- Multi-objective optimization -- Constraint-handling; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2019.05.003
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: Combining simulation models and multi-objective optimization can help solving complex land use allocation problems by considering multiple, often competing demands on landscapes, such as agriculture, (drinking) water provision, or biodiversity conservation. The search for optimal land use allocations has to result in feasible solutions satisfying "real-world" constraints. We here introduce a generic and readily applicable tool to integrate user-specific spatial models (e.g. assessing different ecosystem services) for a Constrained Multi-objective Optimization of Land use Allocation (CoMOLA). The tool can handle basic land use conversion constraints by either a newly and specifically developed method to repair infeasible solutions or by penalizing constraint violation. CoMOLA was systematically tested for different levels of complexity using a virtual landscape and simple ecosystem service and biodiversity models. Our study shows that using repair mechanisms seems to be more effective in exploring the feasible solution space while penalizing constraint violation likely results in infeasible solutions. Highlights: Land use optimization with genetic algorithms is hampered by hard area constraints Penalizing infeasible solutions in the selection process for mating is insufficient Repairing infeasible solutions fosters the exploration of the feasible search space Repairing is computationally demanding, especially for narrow constraints Optimization efficiency decreases strongly with increasing number of spatial units